CAUP Researchers: Carlos J. A. P. Martins
New Constraints on variations of the fine structure constant from CMB anisotropies,
The early universe is a unique laboratory in which to probe fundamental physics, and the Cosmic Microwave Background (CMB) is among its cleanest probes. A number of recent astrophysical studies suggest that some of nature's fundamental couplings (including the fine-structure constant, alpha) are spacetime varying quantities and had different values in the recent universe, at redshifts z~2-3. The CMB temperature and polarization anisotropies are sensitive to variations of alpha and thus allow us model-independent measurements of it in the early universe (at the epoch of recombination, corresponding to a redshift z~1000). Our analysis of the latest available CMB data, together with HST measurements of the Hubble constant improves on previous constraints by a factor of 3, and for the first time pushes these constraints below the percent level .
Nature is characterized by a set of physical laws and fundamental dimensionless couplings which historically have been assumed to be spacetime invariant. For the former this is a cornerstone of the modern scientific method, but for the latter it's a simplifying assumption with no further justification. There is ample experimental evidence showing that fundamental couplings run with energy, and many particle physics and cosmology models suggest that they must also roll in time (and possibly ramble in space). Searching for these is a challenging but powerful probe of fundamental physics, and the early universe is an ideal place to do so.
The astonishing agreement between theoretical expectations and current measurements of the CMB temperature and polarization anisotropies has pushed it to the forefront of these searches. In particular, CMB anisotropies are sensitive to variations of fundamental couplings such as alpha. A time variation of alpha will change the time of recombination and the size of the acoustic horizon at the epoch of photon-electron decoupling, both of which will have characteristic imprints on the CMB angular power spectrum. A key virtue of these measurements is that they probe the value of alpha at this epoch only, and are insensitive to how it may have evolved from that epoch to the present - in other words, they are model-independent.
The CAUP Galaxies & Observational Cosmology group has been responsible for the first astrophysical measurement of alpha using CMB observations . That study also required an analysis of the theoretical underpinnings of these effects and detailed changes to numerical recombination codes (namely RECFAST). Over the last decade we have steadily exploited theoretical, numerical and experimental advances to derive ever-improving constraints on alpha in the early universe. This highlighted paper  is the latest effort in this ongoing project.
Our work, in collaboration with groups in Roma and Paris, uses the latest available CMB results from WMAP, BOOMERanG and CBI but also experiments at arcminute angular scales (ACBAR, QUAD, BICEP) which have a significant impact on the results. Numerical codes have also been improved, and we now carry out a full Markov chain Monte Carlo analysis, using suitably modified COSMOMC/CAMB codes.
Our results are consistent with no variation from the z~1000 epoch to the present day. The relatively weak constraints are due to degeneracies with other cosmological parameters. Notable among these is the Hubble parameter H0, whose change has very similar effect on the CMB to those of alpha. A way to break this degeneracy is to combine CMB data with measurements of H0. By including such data from the HST key project we were able for the first time to obtain better-than-percent-level constraints on alpha, improving on previous results by a factor of 3 (Fig. 1).
These results tightly constrain some models that aim to explain the claimed low-redshift variations. On the other hand, the current CMB data is becoming good enough to allow for studies of correlated variations of several couplings (such as alpha and the proton-to-electron mass ratio), which are expected in most sensible models where one of the couplings varies at all, and our future efforts will be along these lines.
 New constraints on variations of the fine structure constant from CMB anisotropies, Physical Review D 80, 087302 (2009)